Emotional Dissonance, Burnout, and In-Role

Performance Among Nurses and Police Ofcers

Arnold B. Bakker
Erasmus University
Ellen Heuven
Utrecht University

Two studies including 108 nurses and 101 police ofcers tested the proposition that emotionally demanding interactions with recipients may result in
emotional dissonance, which, in turn, may lead to job burnout and impaired
performance. More specically, on the basis of the literature on burnout and
emotional dissonance, the authors hypothesized that emotional job demands
would explain variance in burnout (i.e., exhaustion and cynicism/disengagement) through their inuence on emotional dissonance. In addition, the
authors predicted that emotional dissonance would be (negatively) related to
in-role performance through its relationship with burnout. The ndings of a
series of structural equation modeling analyses supported both hypotheses.
The implications for research and practice are discussed, as well as avenues
for additional research.
Keywords: burnout, emotional demands, emotional dissonance, in-role performance

Nurses and police ofcers seemingly represent opposite poles in the

emotions they are required to express as part of their work role. While nurses
are expected to show a trusting, empathizing concern for their patients, police
ofcers are required to demonstrate a detached, matter-of-fact attitude visa`-vis their clients (civilians). However, despite these divergent work settings,
both professions have two main characteristics in common. First, both nurses
and police ofcers are exposed to emotionally demanding interpersonal
Arnold B. Bakker, Department of Work and Organizational Psychology, Erasmus University, Rotterdam, the Netherlands; Ellen Heuven, Department of Social and Organizational
Psychology, Utrecht University, Utrecht, the Netherlands
Correspondence concerning this article should be addressed to Arnold B. Bakker, Department of Work and Organizational Psychology, Erasmus University Rotterdam, Institute of
Psychology, Woudestein, T12-47, P.O. Box 1738, 3000 DR Rotterdam, the Netherlands.
E-mail: bakker@fsw.eur.nl
423
International Journal of Stress Management
2006, Vol. 13, No. 4, 423 440

Copyright 2006 by the American Psychological Association

1072-5245/06/$12.00 DOI: 10.1037/1072-5245.13.4.423

424

Bakker and Heuven

interactions, such as confrontation with death and illness, violence, and

victims of crime or accidents on a daily basis (e.g., Brown & Campbell, 1994;
Burke, 1994; Schaufeli & Enzmann, 1998). Second, the (non) display of
certain types of emotions is an important task requirement for both nurses and
police ofcers. That is, both professions call for a regulation of feelings and
expressions as part of the work role, which has been referred to as emotion
work (Hochschild, 1983). The central aim of the present study is to examine
how emotion work affects human service professionals well-being and
performance.

OCCUPATIONAL BURNOUT

Burnout has been dened as a specic kind of occupational stress

reaction among human service professionals, resulting from demanding and
emotionally charged interactions with recipients (Maslach, 1982; Maslach &
Schaufeli, 1993). Although several recent studies have shown that burnout is
not restricted to the service professions (e.g., Bakker, Demerouti & Schaufeli,
2002; Demerouti, Bakker, Nachreiner & Schaufeli, 2001; Leiter & Schaufeli,
1996), the syndrome is most prevalent among human services providers
(Schaufeli & Enzmann, 1998).
Two core dimensions of burnout can be distinguished: emotional exhaustion and depersonalization (cf. Maslach, 1982; Cordes & Dougherty,
1993; Demerouti et al., 2001). In response to the chronic emotional strain in
daily interactions with recipients, human service providers may feel emotionally overextended and drained by their interactions with other people
(Leiter & Maslach, 1988). A way of coping with this emotional exhaustion
is to decrease investments in relationships with recipients by emotionally
distancing oneself from them (Maslach, Jackson & Leiter, 1996). Decreased
involvement is expressed by a cynical and dehumanizing attitude toward
recipients, reduced empathy, and by blaming the victim (Schaufeli &
Enzmann, 1998). This detached attitude or depersonalization may vary from
nurses, considering their patients as impersonal objects (that kidney on the
operation table), to police ofcers, blaming a raped woman for walking the
streets alone at night. As a result, human service professionals are unable to
perform adequately, which, in turn, may result in a decline in their feelings
of professional efcacy (Bakker, Schaufeli, Sixma, Bosveld, & Van Dierendonck, 2000; Leiter, 1993). However, several scholars (e.g., Cordes &
Dougherty, 1993; Demerouti et al., 2001; Shirom, 1989) have argued that
reduced professional efcacy is not a genuine burnout component, because it
is only weakly related to emotional exhaustion and depersonalization (or
cynicism) and has different predictors.

Emotional Dissonance and Burnout

425

In the present study, we applied two alternative burnout instruments to

measure burnout among nurses and police ofcers. The Oldenburg Burnout
Inventory (OLBI) (Demerouti, Bakker, Vardakou, & Kantas, 2003) is distinct
from the most frequently used Maslach Burnout Inventory (MBI; Maslach et
al., 1996) in that it covers both negatively and positively framed items,
includes affective but also physical and cognitive aspects, and extends the
concept of depersonalization beyond distancing oneself emotionally from
recipients to work objects and work content.

EMOTIONAL JOB DEMANDS

Specic about the two groups in our present study are the extreme
emotional situations to which nurses and police ofcers are exposed during
their work. Police ofcers are constantly confronted with societys interpersonal violence, confrontational interactions with individuals, and emotionally
charged encounters with victims of crimes and accidents (Brown & Campbell, 1994; Toch, 2002). Likewise, nurses are in frequent, close contact with
severe illness, suffering, and death (Le Blanc, Bakker, Peeters, Van Heesch,
& Schaufeli, 2001). Initially, dealing with these types of severe human
problems in intense, emotionally charged interactions of long duration with
recipients was considered to be the main source of burnout among human
service professionals (Maslach & Jackson, 1984; Pines & Aronson, 1988).
However, contrary to these initial expectations, empirical ndings have
not supported this assumption (Schaufeli & Enzmann, 1998). Job-related
stressors are generally found to more strongly correlated with burnout than
are client-related stressors. Specically, in empirical studies among police
ofcers, organizational constraints such as workload and inadequate systems
of supervision and management were found to be more important for the
development of burnout complaints than frequent interactions with difcult
or aggressive delinquents (Brown & Campbell, 1990; Kop, Euwema, &
Schaufeli, 1999). In a similar vein, burnout complaints among nurses have
not been consistently related to occupation specic stressors, such as confrontation with death and dying and interaction with difcult patients. Nonpatient-related job demands are experienced as even more stressful (see
Schaufeli & Enzmann, 1998, for an overview).

EMOTIONAL DISSONANCE

Emotion work occurs in face-to-face contacts with the public in which

expressions are regulated as part of the job to produce an emotional state in

426

Bakker and Heuven

the recipients or to comply with the organizational rules concerning emotional expression (Grandey, 2000). More specically, emotion work refers to
the psychological efforts necessary to express organizationally desired emotions during interactions with the public (Morris & Feldman, 1997; Zapf,
2002). Police ofcers are expected to manage their emotions to obtain a facial
and physical expression that is neutral, solid, and controlled. Their work
environment asks for a suppression of emotions to adequately deal with, for
example, conict situations, manipulation, and aggression. On the other
hand, police ofcers are also asked to show compassion and understanding
toward, for example, victims of crime. Thus, police ofcers need to master
the art of constantly switching between this human and disciplinary emotional expression. In a similar vein, nurses are required to express a wide
variety of emotions during their interactions with patients. They have to
switch between keeping a certain emotional distance toward their patients to
secure a professional attitude on the one hand, and showing a caring,
compassionate attitude on the other. This is also known as detached concern (Lief & Fox, 1963).
The performance of emotion work may have positive consequences for
employees, such as the facilitation of interpersonal encounters with recipients, task effectiveness, and self-expression (Adelmann, 1995; Ashforth &
Humphrey, 1993). However, several scholars have argued that the regulation
of emotions as part of the work role may be stressful and detrimental to
health. These positive and negative consequences of emotion work can be
traced back in differential effects on stress-related illnesses and burnout
across various empirical studies (see Zapf, 2002, for an overview). However,
consistent and unequivocal relations have been found between a specic
element of emotion work, namely emotional dissonance, and burnout complaints across a wide variety of human service professions (e.g., Abraham,
1998; Brotheridge & Lee, 1998; Heuven & Bakker, 2003; Zapf, Seifert,
Schmutte, Mertini, & Holz, 2001; Zapf, Vogt, Seifert, Mertini, & Isic, 1999).
Emotional dissonance refers to the structural discrepancy between felt
emotions on the one hand and the emotional display that is required and
appropriate in the working context on the other (e.g., Zapf et al., 1999, 2002).
Thus, emotional dissonance is the discrepancy between authentic and displayed emotions as part of the job. Increasing empirical evidence supports the
notion that the suppression of true emotions and feelings has detrimental
effects on health and well-being. For example, the study by Heuven and
Bakker (2003) found that the structural discrepancy between inner feelings
and the positive emotional display in the job of cabin attendants was, more
than quantitative job demands and lack of job control, the main predictor of
burnout complaints. Also, the suppression of experienced emotions in jobs
that demand neutrality in emotional expression, such as correctional ofcers,
has been found to be a source of stress (Rutter & Fielding, 1988). Further-

Emotional Dissonance and Burnout

427

more, the inhibition of emotions and the inability to express negative emotions were strong predictors of physical problems and illnesses (including
high blood pressure and cancer) among employees across various studies (see
Grandey, 2000, for an overview).
Because, by denition, emotional dissonance only occurs among employees interacting with recipients, we believe that the equivocal evidence for
the relationship between emotional job demands and burnout (see previous
paragraph) may be attributable to the fact that emotional dissonance plays a
mediating role in this relationship. That is, emotional demands may particularly lead to burnout through the experience of emotional dissonance.
Nurses and police ofcers are faced with situations, such as death, illness,
and violence that trigger emotional reactions, while their professions require
them to inhibit or suppress the emotions that normally occur in reaction to
these situations. Therefore, we propose that the emotional job demands in
nursing and law enforcement, which require the regulation of emotions, will
result in elevated levels of dissonance between displayed and felt emotions.
Emotional dissonance, in turn, will contribute to feelings of exhaustion and
cynicism.
Hypothesis 1: Emotional job demands are a predictor of burnout (exhaustion and cynicism), through the experience of emotional dissonance.
Thus, we expect that emotional dissonance will play a mediating role in
the relationship between emotional job demands and burnout (see also
Figure 1).

IN-ROLE PERFORMANCE

Emotion work has found its way into the business literature because of
its importance for organizational outcomes. Emotions are important facets of

Figure 1. Hypothetical emotion work model.

428

Bakker and Heuven

the products the service industry sells to clients. Thus, smiling air hostesses, empathic nurses, and neutral police ofcers are of main importance for
the publics perception of these professions and for client satisfaction. For
example, a recent study by Tsai (2001) among sales clerks in retail shoe
stores found that customer satisfaction was highly correlated with the frequency of displaying positive emotions. The study also indicated that employees display of positive emotions increased customers willingness to
return to the store. In addition, several studies have suggested that emotional
dissonance may have important (negative) implications for job satisfaction,
organizational commitment, and turnover intentions (Abraham, 1999; Morris
& Feldman, 1997; Zapf et al., 1999).
Although the relationship between emotional dissonance and organizational outcomes has been substantiated in several studies, the psychological
mechanisms explicating this relation have not been addressed so far. In the
present study, we strive to understand how the relation between emotional
dissonance and one of the most important organizational outcomes, in-role
performance, is mediated by burnout. Several studies have pointed at the
negative implications of emotional dissonance for emotional exhaustion and
depersonalization. In turn, burnout has been related to negative outcomes at
the level of the organization, such as elevated levels of absenteeism and
personnel turnover (see Lee & Ashforth, 1996, for an overview). For example, Aiken, Clarke, Sloane, Sochalski, and Silber (2002) found that 43% of
the nurses who report high burnout levels intend to leave their jobs within a
year compared with 11% of their colleagues without such complaints. Another study among British police ofcers found that 25% of sickness absence
could be attributed specically to stress (Brown & Campbell, 1994).
In the current study, we focused exclusively on in-role performance,
because this variable has been primarily related to job demands, such as
emotional task requirements, whereas job resources are most predictive of
extra role performance (Bakker, Demerouti, & Verbeke, 2004). In-role performance can be dened as the ofcially required outcomes and behaviors
that directly serve organizational objectives (Motowidlo & Van Scotter,
1994). In the case of nurses, this includes primary tasks such as providing
injections, serving meals, and washing patients. For police ofcers, in-role
performance consists of, among other things, street surveillance, putting
suspects under arrest, and responding to alert calls by citizens. Bakker et al.
(2004) found that job demands, including work pressure and emotional
demands, were predictive of in-role performance through their relation with
feelings of exhaustion. We expect a similar relationship in the present study.
That is, we expect nurses and police ofcers who experience a structural
discrepancy between the emotions they need to show as part of the job and
their true feelings to drain their energy resources, and subsequently become

Emotional Dissonance and Burnout

429

cynical and detached toward their recipients and their work. In turn, this will
negatively affect their achievement of organizational objectives.
Hypothesis 2: Emotional dissonance is negatively related to in-role
performance, through the experience of exhaustion and cynicism. That
is, exhaustion and cynicism will play a mediating role in the relationship
between emotional dissonance and in-role performance (see also Figure 1).
METHOD
Participants and Procedure

Two studies were conducted in the Netherlands. The participants were

108 nurses (89% response) working in two departments of a large hospital,
and 101 police ofcers (76% response) working in one department of a police
force. All employees could ll out an electronic questionnaire (published on
a secure website) during work time, in a silent, separate room. A newsletter
and an e-mail from the management to all employees explained the goal of
the study. Anonymity and condentiality of the data was emphasized. The
nurses sample included 21 men (19%) and 87 women (81%). Their age
ranged from 21 to 59 years with an average of 40 years (SD 9.93). Mean
organizational tenure was 11 years (SD 8.15). The police ofcers sample
included 65 men (64%) and 36 women (36%). Their age ranged from 20 to
58 years with an average of 35 years (SD 8.04). Mean organizational
tenure was 5 years (SD 4.96).
Measures

Emotional Demands
Emotional demands were assessed with a scale developed by Van Veldhoven and Meijman (1994) and included four items. Examples are: Is your
work emotionally demanding? and Do the people who you meet through
your work intimidate you? (1 never, 5 always).
Emotional Dissonance
Emotional dissonance was measured following Zapf, Vogt, Seifert, and
Mertinis (1998) conceptualization and operationalization of this construct.

430

Bakker and Heuven

Respondents were asked: How often are you confronted with the following
situations during your work? Four items were used to assess emotional
dissonance. Examples are: Having to show certain feelings to patients
(civilians) that do not correspond with the way you feel at that moment, and
Having to show positive feelings to patients (civilians), while in fact you
feel indifferent (1 never, 5 always).

Burnout
Burnout was assessed with the OLBI (Demerouti et al., 2003) in the
study with nurses and with the Maslach Burnout InventoryGeneral Survey
(MBI-GS; Schaufeli, Leiter, Maslach, & Jackson, 1996) in the study with
police ofcers. The reason for using two different instruments was that we
wanted to generalize our ndings across measurement instruments. Previous
research has indicated that the OLBI and MBI-GS measure conceptually
similar concepts (Demerouti et al., 2003). Both instruments include scales to
assess the two core dimensions of burnout, namely exhaustion and cynicism
(this is called disengagement in the OLBI).
OLBI. The eight items of the exhaustion subscale are generic and refer
to general feelings of emptiness, overtaxing from work, a strong need for rest,
and a state of physical exhaustion. Example items are After my work, I
regularly feel worn out and weary, and After my work, I regularly feel
totally t for my leisure activities (reversed) (1 totally disagree, 4
totally agree). Four items are positively worded and four are worded negatively.
Disengagement refers to distancing oneself from the object and the
content of ones work (including recipients) and to negative, cynical attitudes
and behaviors toward ones work in general. This subscale also comprises
eight items, including I frequently talk about my work in a negative way,
and I get more and more engaged in my work (reversed). Similar answering categories as for exhaustion were used. Again, four items are positively
worded and four are worded negatively.
MBI-GS. Exhaustion is measured with ve items, including I feel
burned out from my work, and I feel tired when I get up in the morning and
have to face another day on the job. Cynicism reects indifference or a
distant attitude toward work and is measured with ve items. Examples are:
I have become less interested in my work since I started this job, and I
have become more cynical about whether my work contributes anything.
Participants were asked to indicate the extent to which they agreed with each
statement using a 7-point rating scale (0 never, 6 every day).

Emotional Dissonance and Burnout

431

In-Role Performance
In-role performance was assessed with three items based on Goodman
and Svyantek (1999). Participants could indicate the extent to which they
found each of the statements characteristic of themselves (0 not at all
characteristic, 6 totally characteristic). An example is: Achieves the
objectives of the job.
Strategy of Analysis
To test our Emotion Work model, we performed structural equation
modeling (SEM) analyses using the AMOS software package (Arbuckle,
1997). The t of the model to the data was examined with the goodness-of-t
index and the root mean square error of approximation (RMSEA). In addition, the non-normed t index, the comparative t index, and the incremental
t index are utilized. In general, models with t indices .90 and an
RMSEA .08 indicate an acceptable t between the model and the data
(Browne & Cudeck, 1989; Hoyle, 1995). All constructs were included as
latent variables in the model. Emotional demands (four items), emotional
dissonance (four items), and in-role performance (three items) were indicated
with the items introduced before. Burnout was indicated with the two
multi-item scales for exhaustion and disengagement/cynicism of the OLBI
(nurses sample) and the MBI-GS (police sample). The hypothesized relationships (see Figure 1) were included in the model.

RESULTS
Descriptive Statistics

Tables 1 and 2 show the means, intercorrelations, and the internal

consistencies (Cronbachs alpha) of the scales included in the analyses. All
Table 1. Means, Standard Deviations, Intercorrelations, and Internal Consistencies
(Cronbachs alpha Between Parentheses) of the Scales Used in the Nurse Study (N 108)
Variable
1.
2.
3.
4.
5.

Emotional demands (15)

Emotional dissonance (15)
ExhaustionOLBI (14)
DisengagementOLBI (14)
In-role performance (06)

SD

2.46
2.24
2.24
2.29
3.59

.46
.36
.45
.43
1.04

(.70)
.40**
.37**
.05
.03

(.64)
.37**
.21*
.08

(.85)
.45**
.26**

(.82)
.22*

(.82)

Note. OLBI Oldenburg Burnout Inventory.

*
p .05. ** p .01.

432

Bakker and Heuven

Table 2. Means, Standard Deviations, Inter-correlations, and Internal Consistencies

(Cronbachs alpha Between Parentheses) of the Scales Used in the Police Study (N 101)
1.
2.
3.
4.
5.

Variable

Emotional demands (15)

Emotional dissonance (15)
ExhaustionMBI-GS (06)
CynicismMBI-GS (06)
In-role performance (06)

2.51
2.53
1.50
1.56
4.27

SD

.71 (.86)
.66
.64**
1.10
.36**
1.06
.32**
1.09 .18

(.76)
.42**
.53*
.22*

(.92)
.53**
.24*

(.80)
.27**

(.80)

Note. MBI-GS Maslach Burnout InventoryGeneral Survey.

p .05. ** p .01.

scales showed good reliabilities, with one exception: the internal consistency
of the emotional dissonance scale was low (.64) in the nurses study. In
addition, several relationships are worth noting from Table 1. First, in both
studies, the raw scores of emotional demands and emotional dissonance were
signicantly and positively related to both burnout dimensions, with one
exception. Unexpectedly, emotional demands was a variable unrelated to
disengagement in the nurses study. Second, of the two emotion variables,
only emotional dissonance was signicantly and negatively related to in-role
performance in one study (police ofcers). Finally, in both studies, the two
burnout dimensions were signicantly and negatively related to in-role performance.
Preliminary analyses revealed that demographic variables were not substantially related to the model components. The only relationships found
indicated that nurses reported somewhat higher levels of cynicism when they
had higher age (r .22, p .05) and when they were more experienced (r
.30, p .01). Regarding the police ofcers, men scored somewhat higher on
emotional demands than women (M 2.70 vs. M 2.21), F(1, 100)
11.61, p .01, and higher on in-role performance (M 4.45 vs. M 3.94),
F(1, 100) 5.17, p .05. However, inclusion of these background variables
in the structural equation model did not signicantly modify the results. They
were therefore omitted from further analyses.

Test of the Proposed Model

To test the proposed model, two separate series of SEM analyses were
performed, one for the nurses and one for the police ofcers. The results
indicated that the hypothesized model tted adequately to the data of both
groups (see second and seventh row in Table 3). Most t indices showed
values of .90 or higher, and the RMSEA was .08. Only the goodness-of-t
index was with .89 (nurses) and .86 (police), somewhat lower than it should
have been. Importantly, all indicators loaded signicantly on the intended

Emotional Dissonance and Burnout

433

Table 3. Results of Structural Equation Modeling Analyses: Fit Indices of Emotion Work
Models for Nurses (N 108) and Police Ofcers (N 101)
Model
Nurses
M1. Proposed model
M2. Alternative model
M3. Alternative model
Null model
Police ofcers
M1. Proposed model
M2. Alternative model
M3. Alternative model
Null model

df

GFI

RMSEA

IFI

CFI

NNFI

90.53
82.45
89.33
414.27

62
60
61
78

.89
.90
.89
.58

.07
.06
.07
.20

.92
.94
.92

.92
.93
.92

.89
.91
.89

108.24
106.68
108.23
643.45

62
60
61
78

.86
.86
.86
.39

.08
.08
.08
.27

.92
.92
.92

.92
.92
.92

.90
.89
.89

Note. GFI goodness-of-t index; RMSEA root mean square error of approximation;
IFI incremental t index; CFI comparative t index; NNFI non-normed t index. M2
alternative model, including the paths from emotional demands to burnout and to performance;
M3 alternative model, including the path from emotional dissonance to performance.

latent factors, and the proposed relationships in the model were signicant,
and in the expected direction (see Figure 2). The coefcient of the path from
emotional demands to emotional dissonance was positive and highly significant (nurses .51, police .73). In addition, the relationship between
emotional dissonance and burnout was positive and signicant (nurses
.44, police .58). Finally, the relationship between burnout and in-role
performance was negative and signicant in both samples (nurses .32,
police .35). The model explained 26% and 53% of the variance in

Figure 2. Standardized solution of the proposed emotion work model (maximum likelihood

estimates) for nurses (N 108; before the slash) and police ofcers (N 101). All parameters
are signicant at the p .05 level.

434

Bakker and Heuven

emotional dissonance, 20% and 34% of the variance in burnout, and 10% and
12% of the variance in in-role performance for nurses and police ofcers,
respectively.
To test the alternative hypothesis that emotional job demands are directly
related to burnout and in-role performance, these two paths were included in
the model for both samples (M2). As can be seen from Table 3, this
alternative and less parsimonious model improved signicantly on the proposed model for the nurses, 2 (2) 8.08, p .05, but not for police
ofcers, 2 (2) 1.56, nonsignicant (NS). Output inspection revealed that
in the nurses sample, only the additional relationship between emotional
demands and burnout was signicant, .33; t 2.77, p .05; the
emotional demandsin-role performance relationship was nonsignicant,
.15; t 1.00. A second alternative model to be tested for both groups
included the direct path from emotional dissonance to in-role performance
(M3). This model was not better than the proposed model (M1) (nurses 2
(1) 1.20, NS), (police ofcers 2 (1) .01, NS), and the additional path
was not signicant (nurses .17, t 1.07, NS; police .02, t 1,
NS).
Taken together, these ndings offer additional support for the proposed
model. In sum, our ndings are consistent with Hypothesis 1: Emotional
demands are related to burnout, mainly through emotional dissonance. Hypothesis 2 is supported as well: Emotional dissonance is only related to
in-role performance through its relationship with burnout. Note, however,
that we only found evidence for real mediation of burnout in the relationship
between emotional dissonance and performance in the police study, because
the correlation between the raw scores of emotional dissonance and in-role
performance was nonsignicant in the nurse study (r .08, NS vs. r
.22; p .05 in the police study).

DISCUSSION

The central aim of the present research among police ofcers and nurses
was to examine: (1) whether emotional job demands are related to burnout
through their inuence on emotional dissonance, and (2) whether emotional
dissonance is related to in-role performance through its inuence on burnout.
Both hypotheses were largely supported by the results of SEM analyses.
First, the emotional dissonance experienced by police ofcers and nurses as
a result of the emotional demands in their jobs provides an important
contribution to explaining why both police ofcers and nurses get emotionally exhausted and detach from their jobs. Thus, our ndings offer a possible
explanation for the lack of empirical evidence with regard to the predictive

Emotional Dissonance and Burnout

435

value of emotional job demands for burnout (Schaufeli & Enzmann, 1998).
These previous ndings seem puzzling in the light of the high prevalence of
burnout among human service professionals as opposed to employees in jobs
that require no client contacts, implying that there is something specic about
the interactions with recipients that may cause burnout complaints. Our
ndings indicate that as a result of the emotionally charged interactions with
patients and civilians, nurses and police ofcers experience a discrepancy
between felt and displayed emotions which, in turn, leads to emotional
exhaustion and cynicism. Thus, the ndings of the present study underscore
the relevance of including emotional dissonance in future studies on the
relationship between emotional demands and job stress in the human services.
Second, emotional dissonance was mainly related to in-role performance
through its relationship with emotional exhaustion and cynicism. These
ndings suggest that nurses and police ofcers who feel a discrepancy
between the emotions they need to show and their true emotions deplete their
energy resources and eventually become cynical toward their recipients and
their work. This may be disastrous for their performance: they may no longer
achieve the objectives of their work, that is, providing high-quality service to
patients and civilians. These ndings are consistent with empirical studies
that tested the inuence of other types of job demands on in-role performance
(e.g., Bakker et al., 2004).
Taken together, our ndings conrm previous theorizing and empirical
research showing that emotion work is a laborious activity (e.g., Hochschild,
1983; Zapf, 2002). Burned-out individuals will logically have to put more
effort into regulating their emotional display compared to their emotionally
vigorous colleagues. That is, the large discrepancy between felt emotions
(i.e., emotional exhaustion) and demanded display (i.e., neutrality or caring,
empathetic expression), reduces the possibilities for a spontaneous expression of emotions and asks for more surface and deep acting. This, in turn, will
further deplete employees emotional resources and result in more burnout
complaints. This reasoning is in line with Hobfolls (1989, 2002) conservation of resources theory, suggesting that people who lack resources are the
most vulnerable to additional losses. Future studies in the eld of emotion
work may formally test such a dynamic model incorporating reciprocal
relationships between emotional dissonance and burnout.
In this context, it may also be interesting to relate the concept of
emotional dissonance to Festingers (1957) theory of cognitive dissonance.
Cognitive dissonance is the incompatibility between two cognitions. The
theory of cognitive dissonance holds that contradicting cognitions serve as a
driving force that compels the mind to acquire or invent new thoughts or
beliefs, or to modify existing beliefs, so as to reduce the amount of dissonance or conict between cognitions. If emotional dissonance evokes the

436

Bakker and Heuven

same psychological mechanism as cognitive dissonance, emotional dissonance can be reduced either by eliminating dissonant emotions, or by adding
new consonant emotions. Thus, police ofcers who have to suppress negative
emotions (e.g., anger, fear) to adequately deal with manipulation and aggression may feel positive emotions (e.g., pride, joy) when they learn that such
an approach facilitates problem solution. Similarly, nurses who cheer up
terminally ill patients may reduce their dissonance when they realize that
their positive emotions alleviate patients pain. These examples clearly show
that reection and performance feedback may be crucial to reduce the
negative impact of emotional dissonance on well-being and performance.
Our ndings also revealed that demographics were not systematically
related to the model variables and did not modify the results of the model
testing. This suggests that the proposed model equally applies for men and
women, nurses and police ofcers, and different age groups. However, we
were not able to compare the mean burnout scores of the two professional
groups, because we used two different burnout instruments. Several scholars
have shown that the prevalence of burnout, as well as the specic patterns of
this syndrome, varies considerably across occupational groups. Nurses generally show elevated levels of emotional exhaustion compared with other
human services professionals, whereas the highest levels of cynicism or
depersonalization are found among police ofcers (Schaufeli & Enzmann,
1998). One explanation for the latter nding is the overrepresentation of men
in law enforcement who, as a rule, experience more cynicism than women.
Another explanation is that the high levels of cynicism among police ofcers
reect their occupational socialization, characterized by objectivity and distance (Schaufeli & Enzmann, 1998). In contrast, several researchers have
argued that the empathic, caring, and understanding attitude required from
nurses may increase their susceptibility to emotional exhaustion (Berger,
1987; Maslach, 1982; Pines & Aronson, 1988). These arguments suggest that
the differences in type of emotional display required as part of the job will
result in differences in type of burnout patterns. That is, the suppression or
nonexpression of emotions among police ofcers may result in relatively
high levels of depersonalization as opposed to the empathic and caring
attitude required from nurses resulting in high levels of emotional exhaustion.
This hypothesis needs additional attention in future research on the relationship between emotion work and burnout.
Finally, an interesting nding in the current study is that both samples
showed identical results even though two distinct burnout measures were
used. Not only does this replication of ndings across two different professional groups provide robust empirical evidence for our emotion work model,
it also conrms the validity of both the MBI and the OLBI as appropriate
instruments to measure burnout (cf. Demerouti et al., 2003).

Emotional Dissonance and Burnout

437

Limitations

Because of the cross-sectional design of the current studies, the postulated relationships between emotional dissonance and burnout dimensions
cannot be interpreted causally. The direction of the causal relationship can
only be determined theoretically. For example, in our study we assumed that
emotional dissonance leads to burnout complaints, while the direction of this
relationship may also be reversed. Indeed, several authors (e.g., Adelmann,
1995; Zapf et al., 1999) have suggested that emotional dissonance can be
considered both as an external organizational demand as well as a stress
reaction that is strongly interrelated with emotional exhaustion. To test such
reciprocal relationships between emotion work and burnout, more elaborated
models are needed using longitudinal research designs or diary studies.
Another limitation of the present research is that both studies were
carried out in the Netherlands. Basically, this calls into question the generalizability of the present ndings to other countries. Moreover, it is likely that
feeling rules strongly vary between different countries. A comparison of
various countries regarding emotion work and its inuence on burnout in
future research would be a valuable contribution to our knowledge in this
domain.

Practical Implications

Despite these limitations, the present ndings have several implications

for nursing and police work, and perhaps more generally for human service
work. Emotional job demands and emotion work are job-specic stressors
that are indissolubly connected with human service work. Nurses will continue to be confronted with physical and emotional suffering, illness, and
death, and dealing with aggression and victims of crime will continue to be
an intrinsic part of police work. Because the results clearly indicate that
emotional dissonance coincides with burnout and reduces in-role performance, it seems important for management to consider ways to reduce (the
impact of) emotional dissonance.
One possibility is to further train nurses and police ofcers in managing
their emotions. Company training with regard to emotion work generally
focuses on teaching employees skills to regulate the emotions of recipients.
Air hostesses are taught how to make passengers feel safe and comfortable
during their ight, nurses learn how to reduce anxiety and dispair, and police
ofcers are instructed how to reduce aggressive behavior and calm victims of
crime. Although the regulation of emotions in recipients may have positive
consequences for the employee involved (e.g., by smoothening the social

438

Bakker and Heuven

interaction), our study underlines the importance of training human service

workers regarding how to deal with their own emotions in a healthy manner.
Employees might be taught how to respond to the emotional demands of their
work while staying true to their own feelings. In addition, the organizational
rules regarding emotional display of frontline employees and the style of
supervision may need revision. Allowing employees to express their true
emotions may seem undesirable for recipients, but the current research
suggests that in the long run this approach may be positive for both clients
and organizations.

REFERENCES
Abraham, R. (1998). Emotional dissonance in organizations: Antecedents, consequences and
moderators. Genetic, Social, and General Psychology Monographs, 124, 229 246.
Abraham, R. (1999). The impact of emotional dissonance on organizational commitment and
intention to turnover. Journal of Psychology, 33, 441 456.
Adelmann, P. K. (1995). Emotional labour as a potential source of job stress. In S. L. Sauter
& L. R. Murphey (Eds.), Organizational risk factors for job stress (pp. 371381).
Washington, DC: American Psychological Association.
Aiken, L. H., Clarke, S. P., Sloane, D. S., Sochalski, J., & Silber, J. H. (2002). Hospital nurse
stafng and patient mortality, nurse burnout, and job dissatisfaction. Journal of the
American Medical Association, 288, 19871993.
Arbuckle, J. L. (1997). Amos users guide. Version 3.6. Chicago: Smallwaters.
Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labour in service roles: The inuence
of identity. Academy of Management Review, 18, 88 115.
Bakker, A. B., Demerouti, E., & Schaufeli, W. B. (2002). Validation of the Maslach Burnout
InventoryGeneral Survey: An Internet study. Anxiety, Stress, and Coping, 15, 245260.
Bakker, A. B., Demerouti, E., & Verbeke, W. (2004). Using the Job DemandsResources
model to predict burnout and performance. Human Resource Management, 43, 83104.
Bakker, A. B., Schaufeli, W. B., Sixma, H., Bosveld, W., & Van Dierendonck, D. (2000).
Patient demands, lack of reciprocity, and burnout: A ve-year longitudinal study among
general practitioners Journal of Organizational Behavior, 21, 425 441.
Berger, D. M. (1987). Clinical empathy. Northville, NJ: Jason Aronson.
Brotheridge, C. M., & Lee, R. T. (1998, August). On the dimensionality of emotional labour:
Development and validation of an emotional labour scale. Paper presented at the rst
conference on Emotions in Organizational Life, San Diego, CA.
Brown, J. M., & Campbell, E. A. (1990). Sources of occupational stress in the police. Work and
Stress, 4, 305318.
Brown, J. M., & Campbell, E. A. (1994). Stress and policing: Sources and strategies.
Chichester: Wiley.
Browne, M. W., & Cudeck, R. (1989). Single sample cross-validation indices for covariance
structures. Multivariate Behavioral Research, 24, 445 455.
Burke, R. J. (1994). Stressful events, work-family conict, coping, psychological burnout, and
well-being among police ofcers. Psychological Reports, 75, 787 800.
Cordes, C., & Dougherty, T. W. (1993). A review and an integration of research on job burnout.
Academy of Management Review, 18, 621 656.

Emotional Dissonance and Burnout

439

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands
resources model of burnout. Journal of Applied Psychology, 86, 499 512.
Demerouti, E., Bakker, A. B., Vardakou, I., & Kantas, A. (2003). The convergent validity of
two burnout instruments: A multitrait-multimethod analysis. European Journal of Psychological Assessment, 19, 1223.
Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row Peterson.
Goodman, S. A., & Svyantek, D. J. (1999). Person-organization t and contextual performance:
Do shared values matter? Journal of Vocational Behavior, 55, 254 275.
Grandey, A. (2000). Emotional regulation in the workplace: A new way to conceptualize
emotional labor. Journal of Occupational Health Psychology, 5, 95110.
Heuven, E. M., & Bakker, A. B. (2003). Emotional dissonance and burnout among cabin
attendants. European Journal of Work and Organizational Psychology, 12, 81100.
Hobfoll, S. E. (1989). Conservation of resources: A new approach at conceptualizing stress.
American Psychologist, 44, 513524.
Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of General
Psychology, 6, 307324.
Hochschild, A. R. (1983). The managed heart. Commercialization of human feeling. Berkeley,
CA: University of California Press.
Hoyle, R. H. (1995). The structural equation modeling approach: Basic concepts and fundamental issues. In R. H. Hoyle (Ed.), Structural equation modeling, concepts, issues, and
applications (pp. 115). Thousand Oaks, CA: Sage.
Kop, N., Euwema, M., & Schaufeli, W. B. (1999). Burnout, job stress and violent behaviour
among Dutch police ofcers. Work and Stress, 13, 326 340.
Le Blanc, P., Bakker, A. B., Peeters, M. C. W., Van Heesch, N. C. A., & Schaufeli, W. B.
(2001). Burnout in oncology care providers: The role of individual differences. Anxiety,
Stress, and Coping, 14, 243263.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the three
dimensions of job burnout. Journal of Applied Psychology, 81, 123133.
Leiter, M. P. (1993). Burnout as a developmental process: Consideration of models. In W. B.
Schaufeli, C. Maslach, & T. Marek (Eds.), Professional burnout: Recent developments in
theory and research (pp. 237250). Washington, DC: Taylor & Francis.
Leiter, M. P., & Maslach, C. (1988). The impact of interpersonal environment of burnout and
organizational commitment. Journal of Organizational Behaviour, 9, 297308.
Leiter, M. P., & Schaufeli, W. B. (1996). Consistency of the burnout construct across
occupations. Anxiety, Stress and Coping, 9, 229 243.
Lief, H. I., & Fox, R. C. (1963). Training for detached concern in medical students. In H. I.
Lief, V. F. Lief & N. R. Lief (Eds.), The psychological basis of medical practice (pp.
1235). New York: Harper & Row.
Maslach, C. (1982). Burnout: The cost of caring. Englewood Cliffs, NJ: Prentice Hall.
Maslach, C., & Jackson, S. E. (1984). Patterns of burnout among a national sample of public
contact workers. Journal of Health and Human Resources Administration, 7, 189 212.
Maslach, C., Jackson, S. E., & Leiter, M. (1996). Maslach Burnout Inventory. Manual (3rd ed.).
Palo Alto, CA: Consulting Psychologists Press.
Maslach, C., & Schaufeli, W. B. (1993). Historical and conceptual development of burnout. In
W. B. Schaufeli, C. Maslach & T. Marek (Eds.), Professional burnout: Recent developments in theory and research (pp. 116). Washington, DC: Taylor & Francis.
Morris, J. A., & Feldman, D. C. (1997). Managing emotions in the workplace. Journal of
Managerial Issues, 9, 257274.
Motowidlo, S. J., & Van Scotter, J. R. (1994). Evidence that task performance should be
distinguished from contextual performance. Journal of Applied Psychology, 79, 475 480.
Pines, A. M., & Aronson, E. (1988). Career burnout: Causes and cures. New York: Free Press.

440

Bakker and Heuven

Rutter, D. R., & Fielding, P. J. (1988). Sources of occupational stress: An examination of

British prison ofcers. Work and Stress, 2, 291299.
Schaufeli, W. B., & Enzmann, D. (1998). The burnout companion to study and practice: A
critical analysis. Washington, DC: Taylor & Francis.
Schaufeli, W. B., Leiter, M. P., Maslach, C., & Jackson, S. E. (1996). The MBI-General Survey.
In C. Maslach, S. E. Jackson, & M. P. Leiter (Eds.), Maslach Burnout Inventory Manual
(3rd ed., pp. 19 26). Palo Alto, CA: Consulting Psychologists Press.
Shirom, A. (1989). Burnout in organizations. In C. L. Cooper & I. T. Robertson (Eds.),
International Review of Industrial and Organizational Psychology (pp. 25 48).
Chichester: Wiley.
Toch, H. (2002). Stress in policing. Washington, DC: American Psychological Association.
Tsai, W. C. (2001). Determinants and consequences of employee displayed positive emotions.
Journal of Management, 27, 497512.
Van Veldhoven, M., & Meijman, T. F. (1994). Het meten van psychosociale arbeidsbelasting
met een vragenlijst: De vragenlijst beleving en beoordeling van de arbeid (VBBA) [The
measurement of psychosocial strain at work: The questionnaire experience and evaluation
of work]. Amsterdam: NIA.
Zapf, D. (2002). Emotion work and psychological strain: A review of the literature and some
conceptual considerations. Human Resource Management Review, 12, 237268.
Zapf, D., Seifert, C., Schmutte, Mertini, H., & Holz, M. (2001). Emotion work and job stressors
and their effects on burnout. Psychology and Health, 16, 527545.
Zapf, D., Vogt, C., Seifert, C., & Mertini, H. (1998). FEMA: Fragebogen zu emotionalen
Arbeitsanforderungen (2nd version) [Questionnaire emotional labour]. Unpublished
manuscript.
Zapf, D., Vogt, C., Seifert, C., Mertini, H., & Isic, A. (1999). Emotion work as a source of
stress: The concept and development of an instrument. European Journal of Work and
Organizational Psychology, 8, 371 400.